Art intelligence: How algorithms are reshaping the auction floor

The convergence of fine art and high technology is becoming increasingly evident in the auction sector, a developing area known as art intelligence. At the New York-based auction house Sotheby’s, this intersection is driven by professionals who apply data analytics to traditional market dynamics.

Kelly Shen, a 2017 graduate of the Massachusetts Institute of Technology, works within this emerging field. Her role involves building algorithms designed to predict art prices by analysing factors such as buying trends and the popularity of specific artists.

In addition to predictive modelling, Shen manages the technical infrastructure required for online auctions. She has worked on cataloguing efforts and ensuring that real-time systems can handle the high volume of traffic generated by thousands of potential bidders visiting the house’s website simultaneously.

Shen’s approach to technology is guided by a pragmatic view of user engagement. While she values elegant algorithms, she notes that sophistication alone is not sufficient if it does not translate into audience interaction. “You can build a super-sophisticated algorithm, but if it doesn’t bring more audience engagement, it doesn’t really matter,” she said.

Her career path was shaped by a double major in computer science and mathematics at MIT, combined with a lifelong passion for drawing. Shen credits a specific course, Project Laboratory in Mathematics, for helping her develop the ability to communicate complex work to diverse audiences. In that class, students presented creative answers to intricate mathematical puzzles.

Beyond her professional work at Sotheby’s, Shen remains active in the academic community. She currently serves as a class officer and vice president of the Association of MIT Alumnae, and volunteers for the MIT Club of New York.